674results about How to "Quick classification" patented technology

A device, method, and system for detecting emboli in the brain is disclosed. A transcranial Doppler photoacoustic device transmits a first energy to a region of interest at an internal site of a subject to produce an image and blood flow velocities of a region of interest by outputting an optical excitation energy to said region of interest and heating said region, causing a transient thermoelastic expansion and produce a wideband ultrasonic emission. Detectors receive the wideband ultrasonic emission and then generate an image of said region of interest from said wideband ultrasonic emission. A Doppler ultrasound signal will also be deployed to image the region of interest. Doppler presents changes in velocity to map blood flow. Additionally, a dye can be given to visualize the brain vasculature and a perfusion measurement can be made in various regions of the brain along with the transcranial Doppler and the photoacoustic screening. Systems are taught using resultory medical data for better triage within an enhanced stroke ecosystem.

It is intended to provide a method whereby a plural number of antibodies against cell surface antigens are quickly classified and to provide a method whereby antigens of the thus classified antibodies are quickly identified. Further, it is intended to provide a method of promoting the utilization of the useful data obtained by the above methods. Furthermore, it is intended to provide an antibody which is effective in treating or diagnosing cancer. Namely, a method of classifying antibodies which comprises: (1) the step of preparing a plural number of antibodies respectively recognizing cell surface antigens; (2) the step of bringing each of these antibodies into contact with a cell of the same species; (3) the step of analyzing each of the cells having been treated in the step (2) by flow cytometry and thus obtaining data indicating the reactivity of each antibody with its cell surface antigen; and (4) the step of comparing the thus obtained data and classifying the individual antibodies depending on the similarity. A method of identifying antigens which further comprises: (5) the step of selecting one to several antibodies from each antibody group formed in the step (4) and identifying antigens thereof; and (6) on the assumption that antigens of the antibodies belonging to a single antibody group are the same or highly related to one another, making relations between the antigens having been identified in the step (5) and the antibody groups to thereby identify the antigens. An antibody against HER1, an antibody against HER2, an antibody against CD46, an antibody against ITGA3, an antibody against ICAM1, an antibody against ALCAM, an antibody against CD147, an antibody against C1qR, an antibody against CD44, an antibody against CD73, an antibody against EpCAM and an antibody against HGFR, each obtained by using the above methods.

A cDNA molecule corresponding to a newly discovered human gene is disclosed. The new gene, which is frequently deleted in liver cancer cells and cell lines, is called the DLC-1 gene. Because the gene is frequently deleted in liver cancer cells, but present in normal cells, it is thought to act as a tumor suppressor. This gene is also frequently deleted in breast and colon cancers, and its expression is decreased or undetectable in many prostate and colon cancers. Also disclosed is the amino acid sequence of the protein encoded by the DLC-1 gene. Methods of using these biological materials in the diagnosis and treatment of hepatocellular cancer, breast cancer, colon cancer, prostate cancer, and adenocarcinomas are presented.

The invention discloses an urban road marker automatic sorting method based on vehicle-mounted laser scanning point cloud. The method includes: S1) a step of subjecting original point cloud data to road surface segmentation based on wheel path data to obtain road surface point cloud data; S2) a step of subjecting the obtained road surface point cloud data to binarization processing and extracting road marker points; S3) a step of clustering the road marker points and separating independent road marker targets; S4) a step of sorting large road markers and small road markers according to the dimensions of the obtained road marker targets; S5) a step of subjecting the large road markers to sorting processing based on wheel paths and road edge lines; and S6) a step of subjecting the small road markers to sorting processing based on deep learning and principal component analysis. The method rapidly and accurately extracts and sorts urban road markers, largely reduces the time and labor cost for data processing, and effectively guarantees traffic safety and intelligent drive reliability.

The invention discloses a face recognition method without face data training. The invention does not need massive face data training, and automatically recognizes the collected face pictures under thecondition of lacking a face base database, which is convenient to quickly collect and classify the massive human data, reduces the workload of manual recognition, and has good adaptability to the recognition task of strangers. The invention realizes the real-time updating of the data in the human face base, improves the definition of the image in the human face base, and improves the accuracy ofthe human face recognition. The invention also discloses a face recognition system without a face base library. The face feature clustering module of the invention makes up for the problem of inaccurate judgment of the side face of the MTCNN, improves the accuracy of the photograph acquisition and improves the accuracy of the face recognition. So that face recognition technology can be used in a wider and more complex scene.

The invention relates to the technical field of conducting on-line detection through the machine vision and image processing technology, mainly relates to a method for conducting on-line detection on the non-continuity lithium battery thin film defects through a machine vision system and provides a non-continuity lithium thin film defect on-line detection method based on the machine vision. A grey level mutational point is obtained through adjacent grey level point difference method of three horizontal scanning lines, and therefore a continuity thin film section is determined; an optimal threshold value algorithm is used for achieving the binarization segmentation of a grey level image, a reversing large area defecting method is used for the binarization image to position a defective target, the geometrical and projection characteristics of the defects are extracted to be used as identification parameters, and finally the minimum Euclidean distance is used for achieving the rapid identification and classification of the defect target.

The invention provides a space domain and frequency domain characteristic based parallel SAR (synthetic aperture radar) image classification method. The method includes: by combining space domain and frequency domain characteristics of SAR images and based on the parallel computation environment, dividing the SAR images into n blocks prior to selecting a small image block in the size of 8*8 pixels around each pixel from each image block, computing corresponding wavelet energy features, gray-level co-occurrence matrix features and filtered gray-level average features of each pixel in each small image block, recovering the wavelet energy features, gray-level co-occurrence matrix features and filtered gray-level average features of each pixel in the n small image blocks to obtain wavelet energy features, gray-level co-occurrence matrix features and filtered gray-level average features of each pixel in the SAR images, forming the features into feature vectors for clustering, and finally classifying the SAR images. According to the method, quick classification of the SAR images depends on efficient information processing capability of a parallel cluster computer system, quick classification is realized, and the problem of low speed of SAR image classification in large data volume is solved.

Systems and methods for inspecting a surface of a specimen such as a semiconductor wafer are provided. A system may include an illumination system configured to direct a first beam of light to a surface of the specimen at an oblique angle of incidence and to direct a second beam of light to a surface of the specimen at a substantially normal angle. The system may also include a collection system configured to collect at least a portion of the first and second beams of light returned from the surface of the specimen. In addition, the system may include a detection system. The detection system may be configured to process the collected portions of the first and second beams of light. In this manner, a presence of defects on the specimen may be detected from the collected portions of the first and second beams of light.

The present invention relates to a system, method, and computer program product for recognition objects in a domain which combines feature-based object classification with efficient search mechanisms based on swarm intelligence. The present invention utilizes a particle swarm optimization (PSO) algorithm and a possibilistic particle swarm optimization algorithm (PPSO), which are effective for optimization of a wide range of functions. PSO searches a multi-dimensional solution space using a population of “software agents” in which each software agent has its own velocity vector. PPSO allows different groups of software agents (i.e., particles) to work together with different temporary search goals that change in different phases of the algorithm. Each agent is a self-contained classifier that interacts and cooperates with other classifier agents to optimize the classifier confidence level. By performing this optimization, the swarm simultaneously finds objects in the scene, determines their size, and optimizes the classifier parameters.

An image focus assessment method is provided that works reliably for images of a variety of relatively dissimilar workpieces or workpiece features. The focus assessment method is based on analysis of a single image (without the benefit of comparison to other images). The robustness of the focus assessment method is enhanced by the use of at least one classifier based on a plurality of focus classification features. In one application, a primary advantage of assessing focus from a single image is that an overall workpiece inspection time may be reduced by avoiding running an autofocus routine if an image is already in focus. In various embodiments, the focus assessment method may include an ensemble of classifiers. The ensemble of classifiers can be trained on different training data (sub)sets or different parameter (sub)sets, and their classification outcomes combined by a voting operation or the like, in order to enhance the overall accuracy and robustness of the focus assessment method.

The invention provides a commodity style classification determination method and an apparatus thereof. The method comprises the following steps of acquiring a commodity picture, and using a trained convolutional neural network to extract a characteristic vector of the commodity picture; calculating cluster density of the characteristic vector, and according to the cluster density, calculating a density distance between the characteristic vector and a first characteristic vector whose cluster density value is greater than a cluster density value of the characteristic vector; according to the cluster density of the characteristic vector and the density distance, determining an initial quantity and an initial center of a characteristic vector cluster; according to the initial quantity and the initial center of the cluster, carrying out characteristic vector clustering on the commodity picture, and acquiring a cluster result of a cluster stabilization condition satisfying setting; and according to the cluster result, determining commodity style classification. In the technical scheme provided in embodiments of the invention, an automatic, rapid, accurate and reliable classification basis is provided for a commodity style, accuracy and efficiency of commodity style classification are increased, and working strength of a worker is reduced.

A process in which a ridge flow analysis of a fingerprint image can be used to determine a fingerprint classification. Areas of like valued ridge flow are collected as regions and said regions are then joined together when edges of the regions exist. These vertices are then reduced to form polygons. Polygons that have a predetermined number of vertices (i.e., 8) and contain values from across a range of the possible ridge flow values (i.e., one of each possible ridge flow value) from the ridge flow table are primary polygons associated with singularity points. These singularity points are further identified as cores or deltas allowing the print image to be globally classified as loop, arch or whorl. Further techniques including, but not limited to, ridge counting can be used to classify the fingerprint as a left or right loop, plain or tented arch, or whorl.

The distribution states of defects are analyzed on the basis of the coordinates of defects detected by an inspection apparatus to classify them into a distribution feature category, or any one of repetitive defect, congestion defect, linear distribution defect, ring / lump distribution defect and random defect. In the manufacturing process for semiconductor substrates, defect distribution states are analyzed on the basis of defect data detected by an inspection apparatus, thereby specifying the cause of defect in apparatus or process.